Changes in This Release for Oracle Database Administrator's Guide

In-Memory Column Store

The In-Memory Column Store (IM column store) in an optional area in the SGA that stores whole tables, table partitions, individual columns, and materialized views in a compressed columnar format. The database uses special techniques to scan columnar data extremely rapidly. The IM column store is a supplement to rather than a replacement for the database buffer cache.

See "Using the In-Memory Column Store".

Data Pump Support for the In-Memory Column Store

Data Pump can keep, override, or drop the In-Memory clause for database objects being imported.

See "Data Pump and the IM Column Store".

Force full database caching mode

To improve performance, you can force an instance to store the database in the buffer cache.

See "Using Force Full Database Caching Mode".

Big Table Cache

The Automatic Big Table Caching feature enables parallel queries to use the buffer cache.

See "Memory Architecture Overview".

Attribute-clustered tables

Attribute clustering specifies a directive for heap-organized tables to store data in close proximity on disk, providing performance and data storage benefits. This directive is only applicable for direct path operations, such was a bulk insert or a move operation.

See "Consider Using Attribute-Clustered Tables".

Zone maps

A zone is a set of contiguous data blocks on disk. A zone map tracks the minimum and maximum of specified columns for all individual zones. The primary benefit of zone maps is I/O reduction for table scans.

See "Consider Using Zone Maps".

Advanced index compression

Advanced index compression results in excellent compression ratios while still providing efficient access to the indexes. Advanced index compression works at the block level to provide the best compression for each block, which means that users do not require knowledge of data characteristics. Advanced index compression automatically chooses the right compression for each block.

See "Creating an Index Using Advanced Index Compression".

Preserving the open mode of PDBs when the CDB restarts

You can preserve the open mode of one or more PDBs when the CDB restarts by using the ALTER PLUGGABLE DATABASE SQL statement with a pdb_save_or_discard_state clause.

See "Preserving or Discarding the Open Mode of PDBs When the CDB Restarts".

The USER_TABLESPACES clause of the CREATE PLUGGABLE DATABASE statement

You can use this clause to separate the data for multiple schemas into different PDBs. For example, when you move a non-CDB to a PDB, and the non-CDB had a number of schemas that each supported different application, you can use this clause to separate the data belonging to each schema into a separate PDB, assuming that each schema used a separate tablespace in the non-CDB.

See "User Tablespaces".

Excluding data when cloning a PDB

The NO DATA clause of the CREATE PLUGGABLE DATABASE statement specifies that a PDB's data model definition is cloned but not the PDB's data.

See "Excluding Data When Cloning a PDB".

Default Oracle Managed Files file system directory or Oracle ASM disk group for a PDB's files

The CREATE_FILE_DEST clause specifies the default location.

See "File Location of the New PDB".

Create a PDB by cloning a non-CDB

You can create a PDB by cloning a non-CDB with a CREATE PLUGGABLE DATABASE statement that includes the FROM clause.

See "Creating a PDB by Cloning an Existing PDB or Non-CDB".

The logging_clause of the CREATE PLUGGABLE DATABASE and ALTER PLUGGABLE DATABASE statement

This clause specifies the logging attribute of the PDB. The logging attribute controls whether certain DML operations are logged in the redo log file (LOGGING) or not (NOLOGGING).

See "PDB Tablespace Logging" for information about this clause and the CREATE PLUGGABLE DATABASE statement. See "Modifying a PDB with the ALTER PLUGGABLE DATABASE Statement" for information about this clause and the ALTER PLUGGABLE DATABASE statement.

The pdb_force_logging_clause of the ALTER PLUGGABLE DATABASE statement

This clause places a PDB into force logging or force nologging mode or takes a PDB out of force logging or force nologging mode.

See "Modifying a PDB with the ALTER PLUGGABLE DATABASE Statement".

The STANDBYS clause of the CREATE PLUGGABLE DATABASE statement

This clause specifies whether the new PDB is included in standby CDBs.

See "PDB Inclusion in Standby CDBs".

Querying user-created tables and views across all PDBs

The CONTAINERS clause enables you to query user-created tables and views across all PDBs in a CDB.

See Querying User-Created Tables and Views Across All PDBs.

Oracle Clusterware support for the Diagnosability Framework

Oracle Clusterware uses the Diagnosability Framework and ADR for recording diagnostic trace data and the Clusterware alert log.

See "ADR in an Oracle Clusterware Environment".

READ object privilege and READ ANY TABLE system privilege

READ privilege on an object enables a user to select from an object without providing the user with any other privileges.

See "System and Object Privileges for External Tables" and Oracle Database Security Guide for more information.

Oracle Multitenant option

Oracle Multitenant option enables an Oracle database to function as a multitenant container database (CDB) that includes one or many customer-created pluggable databases (PDBs). A PDB is a portable collection of schemas, schema objects, and nonschema objects that appears to an Oracle Net client as a non-CDB. All Oracle databases before Oracle Database 12c were non-CDBs. You can unplug a PDB from a CDB and plug it into a different CDB.

See Part VI, "Managing a Multitenant Environment".

Resource Manager support for a multitenant environment

Resource Manager can manage resources on the CDB level and on the PDB level. You can create a CDB resource plan that allocates resources to the entire CDB and to individual PDBs. You can allocate more resources to some PDBs and less to others, or you can specify that all PDBs share resources equally.

See Chapter 44, "Using Oracle Resource Manager for PDBs with SQL*Plus".

Full transportable export/import

Full transportable export/import enables you to move a database from one database instance to another. Transporting a database is much faster than other methods that move a database, such as full database export/import. In addition, you can use full transportable export/import to move a non-CDB (or an Oracle Database 11g Release 2 (11.2.0.3) database) into a PDB that is part of a CDB.

See Chapter 15, "Transporting Data".

New administrative privileges for separation of duties

Oracle Database now provides administrative privileges for tasks related to Oracle Recovery Manager (Oracle RMAN), Oracle Data Guard, and Transparent Data Encryption. Each new administrative privilege grants the minimum required privileges to complete tasks in each area of administration. The new administrative privileges enable you to avoid granting SYSDBA administrative privilege for many common tasks.

See "Administrative Privileges"

Database Smart Flash Cache support for multiple flash devices

A database instance can access and combine multiple flash devices for Database Smart Flash Cache without requiring a volume manager.

See "Database Smart Flash Cache Initialization Parameters".

Temporary undo

Undo for temporary objects is stored in a temporary tablespace, not in the undo tablespace. Using temporary undo reduces the amount of undo stored in the undo tablespace and the size of the redo log. It also enables data manipulation language (DML) operations on temporary tables in a physical standby database with the Oracle Active Data Guard option.

See "Managing Temporary Undo". Also, see Oracle Data Guard Concepts and Administration for information about the benefits of temporary undo in an Oracle Data Guard environment.

Move a data file online

You can move a data file when the data file is online and being accessed. This capability simplifies maintenance operations, such as moving data to a different storage device.

See "Renaming and Relocating Online Data Files".

Multiple indexes on the same set of columns

You can create multiple indexes on the same set of columns to perform application migrations without dropping an existing index and recreating it with different attributes.

See "Understand When to Create Multiple Indexes on the Same Set of Columns".

Move a partition or subpartition online

DML operations can continue to run uninterrupted on a partition or subpartition that is being moved without using online table redefinition.

See "Moving a Table to a New Segment or Tablespace".

Online redefinition of a table in one step

You can use the REDEF_TABLE procedure in the DBMS_REDEFINITION package to perform online redefinition of a table's storage properties in a single call to the procedure.

See "Performing Online Redefinition with the REDEF_TABLE Procedure".

Online redefinition of tables with multiple partitions

To minimize downtime when redefining multiple partitions in a table, you can redefine these partitions online in a single session.

See "Online Redefinition of One or More Partitions".

Online redefinition of tables with Virtual Private Database (VPD) policies

To minimize downtime, tables with VPD policies can be redefined online.

See "Handling Virtual Private Database (VPD) Policies During Online Redefinition".

New time limit parameter in the FINISH_REDEF_TABLE procedure

The dml_lock_timeout parameter in the FINISH_REDEF_TABLE procedure in the DBMS_REDEFINITION package can specify how long the procedure waits for pending DML to commit.

See step 8 in "Performing Online Redefinition with Multiple Procedures in DBMS_REDEFINITION".

Invisible columns

You can make individual table columns invisible. Any generic access of a table does not show the invisible columns in the table.

See "Understand Invisible Columns".

Optimized ALTER TABLE...ADD COLUMN with default value for nullable columns

A nullable column is a column created without using the NOT NULL constraint. For certain types of tables, when adding a nullable column that has a default value, the database can optimize the resource usage and storage requirements for the operation. It does so by storing the default value for the new column as table metadata, avoiding the need to store the value in all existing records.

See "Adding Table Columns".

Copy-on-write cloning of a database with CloneDB

When cloning a database with CloneDB, Oracle Database can create the files in a CloneDB database based on copy-on-write technology, so that only the blocks that are modified in the CloneDB database require additional storage on disk.

See "Cloning a Database with CloneDB".

DDL log

When the logging of DDL statements is enabled, DDL statements are recorded in a separate DDL log instead of the alert log.

See "DDL Log".

Debug log

Some information that can be used to debug a problem is recorded in a separate debug log instead of the alert log.

See "Debug Log".

Full-word options for the Server Control (SRVCTL) utility

For improved usability, each SRVCTL utility option is a full word instead of single letter.

See "SRVCTL Command Reference for Oracle Restart".

Transaction Guard and Application Continuity

Transaction Guard ensures at-most-once execution of transactions to protect applications from duplicate transaction submissions and associated logical errors. Transaction Guard enables Application Continuity, which is the ability to replay transactions after recoverable communication errors.

See "Using Transaction Guard and Application Continuity".

Enhanced statement queuing

Critical statements can bypass the parallel statement queue. You can set the resource plan directive PARALLEL_STMT_CRITICAL to BYPASS_QUEUE for a high-priority consumer group so that parallel statements from the consumer group bypass the parallel statement queue.

See "Creating Resource Plan Directives".

New Job Types

Several new script jobs have been added that permit running custom user scripts using SQL*Plus, the RMAN interpreter, or a command shell for the computer platform.

See "Script Jobs".

The IGNORECASE argument of ORAPWD

To support strong authentication, Oracle recommends that you set IGNORECASE to n or omit IGNORECASE entirely. The default value of this optional ORAPWD argument is n.

See "Creating a Database Password File with ORAPWD" for further information.

Oracle Restart

Oracle Restart is a feature provided as part of Oracle Grid Infrastructure. Oracle Restart monitors and can restart Oracle Database instances, Oracle Net listeners, and Oracle ASM instances. Oracle Restart is currently restricted to manage single instance Oracle databases and Oracle ASM instances only, and is subject to desupport in future releases. Oracle continues to provide Oracle ASM as part of the Oracle Grid Infrastructure installation for Standalone and Cluster deployments.

For more information about Oracle Restart, see Chapter 4, "Configuring Automatic Restart of an Oracle Database".

For more information about the Oracle Restart deprecation announcement and its replacement, see My Oracle Support Note 1584742.1 at the following URL:

https://support.oracle.com/CSP/main/article?cmd=show&type=NOT&id=1584742.1

Single-character options with Server Control (SRVCTL) utility commands

All SRVCTL commands have been enhanced to accept full-word options instead of the single-letter options. All new SRVCTL command options added in this release support full-word options only and do not have single-letter equivalents. The use of single-character options with SRVCTL commands might be desupported in a future release.

See "SRVCTL Command Reference for Oracle Restart" for further information.

The FILE_MAPPING initialization parameter

The FILE_MAPPING initialization parameter is deprecated. It is still supported for backward compatibility.

See Oracle Database Reference for information about the FILE_MAPPING initialization parameter.

*_SCHEDULER_CREDENTIALS

This view continues to be available for backward compatibility.

See "Specifying Scheduler Job Credentials" for further information.